Machine learning uses statistical techniques for teaching computers with data to perform specific tasks without being explicitly programmed to do so. The goal of machine learning is to construct algorithms that can learn from and make predictions on data. These algorithms work by creating mathematical models which can classify data. The process of creating the models can involve training and fine tuning the model parameters using input data.